What is Python?
Discover what Python really is: how CPython runs your code, the new 3.15 features, where the language shines, and why Meta, Netflix and Dropbox bet on it
Here is a true fact from June 2026. Meta published a blog post celebrating ten years as the main sponsor of the Python Software Foundation, and in that post its engineers wrote something surprising: Python is the most used programming language at Meta. The same company that runs Instagram and Threads relies on Python for the backends of its biggest consumer products, and even the PyTorch machine learning framework that powers most modern AI was born there on top of Python.
If you open any recent tech job posting, Python appears again and again: data analysis, automation, AI, web backends. The TIOBE index, which tracks how often programming languages are searched for, put Python in first place in October 2026 with a rating of 19.23%, far ahead of C and C++. A beginner-friendly language born in 1991 now sits at the center of the industry.
But saying Python is popular does not tell you what it is. That is what this lesson answers. You will learn how Python actually executes your code, why its design philosophy makes it unusually readable, which release is current in November 2026, and the situations where Python is the right tool and the ones where you should reach for something else.
By the end you will write and run your first Python statements, and you will be ready for the next lesson, which walks through installing Python on your own machine.
Key terms
- A program that reads code and executes it directly, instead of turning it into a separate machine-file first.
- The reference implementation of Python, written in C, that most people install; when this course says Python it means CPython.
- Read, Eval, Print, Loop. A live session where you type one statement and immediately see its result.
- A compact set of internal instructions that CPython compiles your source code into before executing it.
- The Python Package Index, the public catalogue where third-party Python packages are published and retrieved.
- Python Enhancement Proposal. A numbered design document, usually naming how a language feature or standard works.
Interpreter
CPython
REPL
Bytecode
PyPI
PEP
1. What Python Actually Is
Python is a high-level, dynamically typed programming language. Break that down:
- High-level means you work with concepts such as lists, dictionaries and strings, and you never manage raw memory addresses the way C programmers do.
- Dynamically typed means a variable can hold a number now and a text string later, and the interpreter figures types out as the program runs.
- One idea per sentence keeps the language teachable: the interpreter reads your file, executes each statement, and gives an error the moment something is wrong.
There is one subtlety worth a diagram. People often repeat the slogan that Python is "interpreted, so there is no compile step". That slogan is wrong in a useful way. CPython does compile your code. It does so transparently, at run time, and the compiler output is not machine code for a specific chip: it is bytecode, a compact set of internal instructions that a virtual machine then executes.
The pipeline below is the mental model you should keep. Everything you write travels from your editor through these stages into your hardware.
Description
A source file passes through CPython's tokenizer, parser and abstract syntax tree, becomes bytecode, and is executed by the Python virtual machine on the CPU and memory.
Read the diagram in three chunks. The blue node is your file on disk. The green
nodes are CPython working quietly at run time: tokens, a parsed tree, bytecode.
The yellow node is the hardware that finally executes the virtual machine's
instructions. When Python reports a SyntaxError, the failure happened at the
parse step, before any bytecode existed. Once you know which stage produced an
error, you know where to look.
📚 Did you know? The language is named after the BBC comedy group Monty Python, not the snake. Their sketches are embedded everywhere: the built-in documentation uses "spam" and "eggs" as placeholder names, and the official Python tutorials historically kept the joking tone.
2. The Zen of Python
Every design decision in Python traces back to one short document: PEP 20, better known as the Zen of Python, written by Tim Peters. It is only nineteen lines, and you can read it from inside Python itself, because it ships as an easter egg:
import thisRun that in the interpreter and the whole poem prints. Among its principles:
- Beautiful is better than ugly.
- Simple is better than complex.
- Readability counts.
- There should be one, and preferably only one, obvious way to do it.
This is why Python reads almost like English. Look at the original "Hello, world" program in several languages and the pattern will be obvious:
# Python reads like a sentence: do this thing, with these pieces
print("Hello, Nuniversity!")
print(2 + 2)
print("Python", 3, "rocks!")The same three lines require no imports, no class wrapper, no main function, and no type declarations. Compare that with the setup a C++ or Java program needs before printing. Python removes ceremony so the idea is the code. For a beginner this is one of the biggest deal-breakers in language choice: you can express a thought the moment you have it.
Nothing in Zen means Python forgives sloppiness. Readability counts because other people will read your code. The course keeps asking you to name things clearly and to keep functions small, and both habits come straight from PEP 20.
3. Popularity You Can Measure
"Popular" needs evidence, so here is the measured kind. The TIOBE index tracks search-engine interest in language names across many engines. In its October 2026 edition Python led with a rating of 19.23%, more than one and a half times C, the runner-up.
Description
A bar chart of TIOBE index ratings in October 2026: Python leads at 19.23 percent, followed by C at 11.33, C++ at 8.87, Java at 8.41, C sharp at 3.79 and JavaScript at 2.77.
Two extra data points make that lead meaningful:
- Python has won TIOBE's "Language of the Year" award six times, more than any other language: 2007, 2010, 2018, 2020, 2021 and 2024.
- A single language series has now held the number-one position for years, with no sign of another language taking it back in late 2026.
Popularity is not only about hype. It buys you something practical: almost any question you will ever ask has been asked before, answered on a forum, or solved by a published package on PyPI, the Python Package Index. That catalogue holds an enormous number of third-party packages, which is why Python programmers spend their time on their problem instead of rebuilding tools.
📚 Did you know? TIOBE publishes a "Hall of Fame" for languages that rise the most within a year, and Python has won it six times: 2007, 2010, 2018, 2020, 2021 and 2024. No other language has more than three titles.
4. Python Releases: the October Rhythm
Python lands a new series every October, and each series keeps receiving fixes for years. Knowing which series you are on matters, because every October the oldest branch loses support. Here is the state on 4 November 2026.
Release | Status as of Nov 2026 | First release | Support ends |
|---|---|---|---|
Python 3.16 | pre-release, planned | 2026-10-06 target | 2032-10 |
Python 3.15 | current stable, bugfix | 2026-10-09 | 2031-10 |
Python 3.14 | bugfix, previous stable | 2025-10-07 | 2030-10 |
Python 3.13 | security-only fixes | 2024-10-07 | 2029-10 |
Python 3.12 | security-only fixes | 2023-10-02 | 2028-10 |
Python 3.10 | reached end of life on 2026-10-01 | 2021-10-04 | 2026-10 |
(Sources: python.org downloads page and devguide.python.org, checked November 2026.)
The lifetimes are longer than they look. The gantt chart shows how the phases overlap: a series stops receiving new features, but the fixes keep flowing for years afterwards.
Description
A Gantt chart of CPython release support: 3.15 stays in bugfix support until October 2028 and security fixes until October 2031, while 3.10 reaches end of life on 1 October 2026.
2026 Updates (as of November 2026)
What changed very recently, and why older material looks different. Every line below was checked against a primary source in November 2026:
- Python 3.15.2 is the newest stable release (3.15.0 on 2026-10-09, 3.15.2 on 2026-10-25). Any tutorial describing Python 3.13 or 3.14 as "the latest" is already slightly out of date.
- Python 3.10 reached end of life on 1 October 2026. Machines still running 3.10 get no fixes; a beginner should not start there.
- Lazy imports (PEP 810) let a new
lazykeyword defer a module's loading until its name is first used, which speeds up program start for applications that import a lot of modules. frozendict(PEP 814) is now a built-in type: an immutable, hashable dictionary, useful for fixed configuration data.sentinel(PEP 661) is a built-in way to create a unique "no value" marker, replacing the hand-rolledobject()idiom.- UTF-8 became the default text encoding (PEP 686): opening a text file without naming an encoding now reads and writes UTF-8 across platforms.
- Unpacking inside comprehensions (PEP 798) simplifies flattening, so
[*row for row in rows]now works. - Free-threaded builds, without the Global Interpreter Lock, are officially supported as of Python 3.14 (PEP 779); 3.15 improved their performance further and added a stable ABI for them (PEP 803).
What actually changed in 3.15, one screen at a time
Description
A mindmap of Python 3.15 changes: faster startup, better built-ins such as frozendict and sentinel, friendlier error messages, UTF-8 text handling and new profiling tools.
The error-message improvement is worth one real example, because it shows the design philosophy working on you, the beginner. Say you are typing quickly and reach for a JavaScript habit:
>>> nums = [1, 2, 3]
>>> nums.push(4)
Traceback (most recent call last):
AttributeError: 'list' object has no attribute 'push'.
Did you mean '.append'?CPython now detects that you probably learned push in another language and
suggests the Python equivalent, .append. Small touches like this are why
people describe learning Python as being helped along by the tool itself.
5. Where Python Fits, and Where It Does Not
The right language is a match with the job, not a fan debate. Python's design costs real time at run: high-level dynamic code runs slower than compiled C or C++, often dramatically slower for tight numeric loops written in pure Python. The skill is knowing which jobs still want it.
Job | Python as the main tool? | Why |
|---|---|---|
Automating a nightly report | yes | great standard library, fast to write |
Cleaning and analysing CSV data | yes | pandas and NumPy do heavy lifting in C |
Building a backend API | yes | Django and FastAPI are mature |
Training or serving ML models | yes, with helpers | PyTorch, scikit-learn, NumPy hold the hot path |
Compilers, game engines, kernels | mostly no | C, C++ or Rust win on latency and control |
Browser frontends | no | JavaScript runs there natively |
Sub-millisecond trading loops | mostly no | latency and memory control matter more than speed of writing |
Nothing on that table is dogma: Python can do all of it with extra work, and the table exists to teach where the effortless path lives.
Comparative Verdict: is Python still the right first language in 2026
- Versus JavaScript: JavaScript is the only choice if your goal is a browser, since it runs there natively. For everything else, Python's cleaner syntax and its data and AI libraries make it the smoother first language on a laptop.
- Versus Java or C#: those languages teach you types, classes and professional structure earlier, which helps later but slows the start. Python lets you think about problems first, then introduces typing when the course needs it.
- Versus C or C++: these teach what the machine really does, at the cost of memory bugs and ceremony. Learning them after Python gives you a mental model to attach that new depth to.
- Verdict: for a first language in late 2026 with no specific platform target, Python is the strongest default, with JavaScript reserved for web frontend goals.
Now a quick check that this mental model is solid. Sort these tasks by which bucket they belong in. Each item describes work a real team has somewhere in a year.
How to answer
Sorting exercise. Place each item into the correct bucket, then check your answer.
Sort each task by how well Python fits it
Drag every card into the bucket where it belongs most naturally.
Items, click an item, then click a bucket to place it:
Great fit for Python
Empty, click to place the selected item
Reach for another tool
Empty, click to place the selected item
The next interactive turns those same ideas into matching pairs, the fastest vocabulary exercise in the course: match each term to what it actually does.
How to answer
Matching exercise with 5 pairs. Match each item on the left with its answer on the right.
Match each term to its job.
Drag answers from here:
6. Real People Using Python Right Now
Abstract arguments for a language matter less than seeing real deployments. Here are three, all pulled from the companies' own engineering posts, and each shows a different reason to be glad the language is easy to read.
Meta, Instagram and Threads. Meta's engineering blog of 30 June 2026 states plainly that Python is its most used programming language, and that Instagram and Threads backends run on it. The post also says Meta funds the language's development through the Python Software Foundation for the tenth year in a row. That means a language designed for readability now carries one of the world's largest social platforms.
Netflix, streaming at planetary scale. Netflix's own technical blog describes Python across "the full content lifecycle": choosing what shows to fund, managing the Open Connect network of CDN appliances that actually streams the video, and running demand-failover and encoding tools. When engineers say Python is their automation glue, Netflix is what that looks like at scale.
Dropbox, one giant migration. Dropbox's desktop client was largely written in Python, and its engineers published the plan they used to move every install from Python 2 to Python 3, finishing the rollout in about seven months. A later post describes type checking nearly 4 million lines of Python with mypy. Their experience shows both faces of the language: quick to build in, and able to learn good tooling as code bases grow.
Python in science and education
Two academic courses picked Python for foundational teaching, and both are free to study: MIT's 6.00.1x and Harvard's CS50. The MIT open course describes Python as "aimed at students with little or no programming experience", and CS50 uses it to teach algorithmic thinking alongside the language. That is a quiet vote of confidence: when the point is learning to think, teachers choose the syntax that disappears, and that points at Python again.
Popularity is a learning advantage, not a personal merit badge. The practical reason to pick a popular language is that every problem you will face has a solved package, a settled answer, or a book already written. Keep those tools in your back pocket while you learn the language itself.
Timing matters too: with 3.10 out of support and 3.15 fresh out in October 2026, there has never been a better moment to start on the current branch. The release you install on day one is the same one the industry is migrating to, instead of a language two decades behind.
Practice Questions
Answer these before moving to the next lesson. Every question assumes only what was written here.
How to answer
Multiple-choice question with 3 options. Select one option, then choose Check Answer.
A teammate claims Python has no compile step because it is an interpreted language. What does CPython actually do before running a script?
How to answer
Multiple-choice question with 5 options. Select one option, then choose Check Answer.
Which release is the current stable Python series as of November 2026?
How to answer
Multiple-choice question with 2 options. Select one option, then choose Check Answer.
Python 3.10 reached end of life on 1 October 2026, so new Linux servers should not be set up with it.
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
A heavy data-analysing program imports many modules but typically uses only a few of them per session. Which Python 3.15 feature is aimed directly at faster startup in that situation?
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
You open a terminal, run `python`, and type `2 + 2`, then press Enter. What happens?
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
Which company stated in a June 2026 engineering post that Python is its most used programming language?
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
Which TIOBE figure correctly describes Python in October 2026?
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
In Python 3.15, what happens if you open a text file for writing without naming an encoding, such as `open('notes.txt', 'w')`?
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
Which job is the *worst* fit for pure, high-level Python code with no extra libraries?
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
A teammate checks their laptop with `python --version` and sees `Python 3.13.12`. What is the most accurate statement about their interpreter in November 2026?
Try It Yourself: your first statements
There is one extra interactive block here that ties all those ideas together. It is not a test: fill the blanks and see whether the pieces line up.
First, look at what the loop itself looks like when you are inside it. The sequence is short and the timing is the whole point: nothing waits for a build or a save.
Description
A sequence diagram of the Python REPL: you type an expression, Python reads and evaluates it, prints the result, and waits for the next statement.
How to answer
Fill-in-the-blank exercise with 2 blanks. Type the missing words, then check your answer.
Complete your first Python session. One blank is a built-in function name, the other starts a comment.
blank("Hello, Python!") # blank starts a comment in Python
Word Bank, drag words into the blanks above:
Two ideas matter here:
printis one of Python's built-in functions, always available with no import.- Comments start at
#and run to the end of the line, which is the cheapest documentation you will ever write.
One last question before the wrap-up, because the process of running code is the thing beginners find most alien at first.
How to answer
Multiple-choice question with 4 options. Select one option, then choose Check Answer.
You are debugging a teammate's script and want the fastest way to test a single line of arithmetic without creating any file. What should you reach for?
Real-World Case Studies
Each case below is short on purpose; the aim is to lodge one image per case: the kind of work and the kind of source claiming it.
Case 1: Meta, June 2026, the tenth sponsorship year
The primary source is Meta's engineering blog post, published 30 June 2026, which states Python is Meta's most used language, specifically points to the Instagram and Threads backends, and notes the company's tenth consecutive year sponsoring the Python Software Foundation. It also describes PyTorch as born at Meta. One image to keep: the readability-first language is carrying planet-scale consumer traffic, and its own development is paid for by the companies that depend on it.
Case 2: Netflix, streaming operations in Python
The primary source is the Netflix Tech Blog, post dated 29 April 2019, which describes Python "through the full content lifecycle", including the Open Connect CDN whose devices are mostly managed by Python applications, the demand-failover tooling, encoding, and personalization work. One image: large parts of the machine behind the play button are gluing and automating systems, exactly where Python's design strengths live.
Case 3: Dropbox, one big migration and four million type-checked lines
The primary sources are two posts: dropbox.tech, 25 September 2018, on completing a company-wide Python 2 to Python 3 rollout in about seven months, and dropbox.tech, 5 September 2019, on type checking close to 4 million lines. One image: an application you probably use today runs on Python, and the team grew into static-checking tooling gradually, meaning your path does not need to start with heavy tooling on day one.
📚 Did you know? Instagram's engineers say on the Django project's own site that Instagram runs the world's largest deployment of Django, a web framework written entirely in Python. Millions of lines of Python get merged and deployed to production many times a day at that scale.
What Comes Next
You have met the language: what CPython is, its philosophy, its release rhythm, who uses it and the shape of the jobs it wins. The next lesson is the practical one. It walks through installing Python on your own machine, choosing an editor, and running both a script and the REPL for real, on your keyboard rather than in theory.
If you want one long, free bibliography to keep beside the course, the freeCodeCamp full-course video below is the same teaching material you have probably already heard about, recorded as a full university-length Python course with exercises.
Description
A full university Python course by Dr. Chuck, useful as parallel homework; it covers the same beginner ground at a slower pace.
Dig deeper
- Python for Everybody, full coursefreeCodeCamp (Dr. Chuck)840 min
- CS50: Introduction to Computer ScienceHarvard via YouTube
- MIT 6.00.1x, Introduction to Computer Science and Programming Using PythonMIT on edX
- The official Python tutorialPython 3.15 documentation
- The Zen of Python (PEP 20)PEP index
- CPython, the reference implementation sourceGitHub
Additional verified references if you want to check the course's claims yourself: the Python release status table, the CPython developer guide, the 3.15 release notes, the TIOBE index, Meta's June 2026 engineering post, the Netflix tech blog and Dropbox's migration write-up from dropbox.tech.
Key Takeaways:
- Python is a high-level, dynamically typed language whose reference implementation, CPython, compiles your source to bytecode and runs it on a virtual machine, so the classic slogan "interpreted means no compile step" is misleading.
- The Zen of Python, PEP 20, is the design north star: readable, simple, consistent. That philosophy is why the language appeals to beginners and to large teams at once.
- In November 2026 the stable Python is 3.15 (3.15.2 on 2026-10-25), with 3.16 pre-released. Python 3.10 lost support on 2026-10-01, and a new series arrives each October, so version hygiene matters from day one.
- Python 3.15 adds lazy imports (PEP 810), the
frozendictandsentinelbuilt-ins, UTF-8 as the default encoding, unpacking inside comprehensions, aprofilingpackage, and friendly error messages such as suggesting.appendwhen you write.push. - Popularity is real and measurable: TIOBE rated Python first place at 19.23% in October 2026, and it has six Language of the Year titles.
- Meta calls Python its most used language, Netflix uses it across the content lifecycle, and Dropbox completed one of the largest 2-to-3 migrations; keep their posts as your go-to case studies.
- Python's sweet spots are automation, data and AI work and web backends; browser frontends, native mobile widgets and microsecond-latency loops are jobs where another tool usually wins.
- The REPL is your always-on experiment loop: one statement in, one answer out, no files created.
- PyPI is the huge third-party catalogue behind "batteries included", so most real work becomes combining packages rather than writing primitives.
- Next lesson installs Python on your machine and gets a script and a REPL session running for real.
References
Every factual claim links to its source, with the access date printed so the source can be rechecked.
Index table — Python first at 19.23%, ahead of C at 11.33%.
Principles list — The nineteen aphorisms quoted in this lesson.
Release status table — Active branches, the October release rhythm and support windows.
Highlights — Lazy imports, frozendict, sentinel, UTF-8 default, comprehension unpacking.
Python across the full content production lifecycle, Open Connect CDN included.